Measures of Skewness

and Kurtosis

To study about a distribution completely, the measures of central tendency

and variation alone do not help. The measures of central tendency
determine the central point of distribution and measures of variation study
the extend to which the items of distribution vary from one another and
from the central value. But they do not study whether a distribution is
symmetrical or not(Skewness is a measure which help us to understanda
distribution.)
Measures of skewness tell us about the direction and the extent of
symmetry or asymmetry in a distribution. In other words, it describes the
shape of the distribution
If a distribution is not symmetrical, we say that it is skewed. In a
perfect symmetrical distribution, mean, median, and mode coincide.
If the frequency curve has a long tail to the right, we say that it is
skewed to the right. This means that mean is greater than the mode. So the
distribution is positively skewed.
Ifthe frequency curve has a long tail to the left it is said to be skewed
to the left and mean is less than the mode. So the distribution is negatively
skewed
Symmetrical

/ Mean Mode Mode Mean

AN Mode Mean

MPORTANT MEASURES OF SKEWNESS

Karl Pearson's coefficient of skewness
Bowley's coefficient of skewness
MSK 2 Business Statistics and Operatlons Research

Karl Pearson's Coefflclent of Skewness

Pearson's coeflicient ofskewness is based upon the differenee
ard deviationoto gvea
mean and mode. This difference is divided by standard
relative measure.
helween
Coefticient of skewness-
Mean-Mode
S.D
This value usually lies between + I and-1.
When the mode is ill-defined we use the empirical
formula
Mode 3 Median-2 Mean.
Therefore,
Coefficient of skewness = Mean-(3 Median -2 Mean)
S.D.
3(Mean - Median)

S.D.
Bowley's Coefficient of Skewness
Bowley's coefficient of skewness is based on
quartiles.
Sk,+-2 Median
This is also called -
between+ 1and-1.
as
quartile measure of skewness and it varies

WORKED EXAMPLES
Example Calculate Karl Pearson's
1
data given below: coefficient of skewness from the
Size (x)
23 4 5
Frequency ()|1018 30
Solution 25 12 3
2

10
d=x-4
10
fd f
2 18 -3 30
36 90
3 2 - 36
30 90 72
- 1
4
25 100 30 30
5 12 0
60
6 3 12
18 12
7 2
2 6
14 12
N 100 328 6 18
-72 234
 Measures of Skewness and Kurtosis
MSK-3

2f
Mean =EA=SZ=3.28
N100
cince mean is in iraction we take deviations from assumed mean

d=
6 N E 234(72
N
100 100
=
2.340.5 18 =
1.35
Bv
B inspection,
the value of mode is
3 (because its corresponding
frequency 1s maximum).

Coefficient ofskewness=Mean-Mode
S.D
3.28-30.28
30.027
1.35 1.35
CHDle 2 Calculate Pearson's coefficient of skewness:
12.5
| 22.5 27.5 32.5 37.5 42.5 47.5
17.5

28 42 54 108 129 61 4533

Solution

d -27.5
X

12.5 28 3 84 252
n.5 42 2 84 168
22.5 54 54 54
27.5 108
32.5 129 0 129 129
37.5 61 2 122 244
42.5 45 3 135 405
47.5 33 4 132 528
500 296 1780
A=27.5, i=5, N=500, 2fd 296, Efa
= =1780

Mean =A+2Ja
N
296
27.5+ x5
500
= 27.5+2.96 = 30.46
 R e s e a r c h asures of Skewness
ness and Kurtosis MSK5

method, modal class is

O p e r a t l o n s

and spection
S t a t i s t i c s

By 40-60
ModeL+ A+
Business

MSK-4
the variabi

value ofmode
- 32.5 (since 2.5 -xi a -25
Mode the
method,

By
spection

frequency
129) A =27-17
has the
maximum

40+ 2 -x20
2+10
40+3.333 =43.333
x5 =8.96
290x5=3.56-0.3505
Coefficient of skewness= Mean -Mode
1780
500 S.D.
500
M e a n - M o d e

Coefficient ofskewness
SS.D.
I
30.46-32.5--2.04 -0.228 Mple 44
50-43.355-0,2635
25.3
From the following data, calculate Karl Pearson's coefficient
Erample
8.96
8.96 ofskewness:
from the following date
lata.
skewness
of
Example 3
Calculate
coefficient

40-60 60-80 80-100

Marks more than
than 10 20 30 40 50 60 70 80
20-40 No. ofstudents 150 140 100 80 80 70 30 14 0
0-20 19 16
Daily expenditure 1 3 25 27

No. of families Solufion

cumulative frequency distribution is given, we convert it to

a
Solution fNo. offamilies Cince the
frequency distribution.
r Daily expenditure
Midx d=
X-50 MarksS+No. ofstudents
X 20
26 52 Midx d5 cf
10 10
0-20 13 25 25 -40 160 10
25 30 0-10 10 5
20-40 360 50
120
40-60 22 50 40 15
9
10-20 40 8 70
1 19 20 25
60-80 19 70
64
20-30 0 0 70
2 32 .0 35
80-100 16 90 30-40 0 80
160 40-50 10 45 120
40 40
| 100
50-60 40 55 64 136
=160 32
A50,i=20,N=100, Zfd =0, 2fa" 60-70 16 5
3 42 126 150
70-80 14 75 830
Mean, = A+ i=50+x20=50
100 150
86
N 830
fd= -86, Lfa'
=

N= 150, 2
A=45, i= 10,

Mean =
A+ xi
160 x20v1.6x20=1.265x20 20 =25.3 45+x10
=
45-5.733
=39.267

100100 =
45 150
 Measures of Skewness and
and
Operations
Research
KurtosisMSK"
MSK-6 Business Statistics
nalysis table
Column Class in which
mode is
no. 10-20 expected to be
20-30 50-60
3086xi0=V5.53-0.329 x10 II
150150 III
=2.2812x10= 22.812
inspection method V
we cannot find the modal value by
From the table,
table. V
do we prepare grouping and analysis
Grouping table
VI
4

0-10 10 (60 This is bimodal series and hence for calculating skewness, we use the
10-20 40) formula,

Skewness =* 3(Mean-Median)
20-30 20 S.D.
20 30

30-40
Median Size of item
10
50
Size of50 item

40-50 10 75th item

66
Median class= 40-50
50-60 56
60-70 16 xi
30 Median =L+
70-80 14 =40+ 10
10
45
 Research ISures of
MSK8 Business Statistics and
Operations Skewness and Kurtosis MSK°
K-9
A85,i=
3(Mean-Median)
10, N=300, fd
Skewness
=778, Lfa =3510
S.D.
3(39.267-45) Mean =A+2Ja
22.81
17.199 854278
x10
22.81 300
= -0.754 =85+25.93
Calculate Karl Pearson's
coefticient of skewness from the
he =110.93
Example 5
data given below:
f (fa
40-50 50-60 60-70 70-80 80-90
90-100| N N Xi
Weekly wages (Rs.)|
No.of workers 5 6 8 | 10 25 30
3510 (778
110-120 120-130 130-140 300300 x10=V11.7-6.725x10 22.3
Weekly wages (Rs.) 100-110
36 50 60 70
No. ofworkers Median Size item
Solution
(2
Since the distribution is skewed (i.e.) it is not symmetrical it would be 300
= Size of I tem=150th item
better to calculate coefficient of skewness based on mean and median.

3(Mean-Median) Median class =110-120

Skewness =
S.D. N
-f
Median=L+2 xi
xWeeklywages
SNo. of workers
= 150-12010 115
Midx d=- cf =110+6 116
10
40-50 45 -20 80 3(Mean- Median)
50-60 55 - 18 54 11 Skewness
S.D.
60-70 8 65 2 - 16 32 19
3(110.93-1l6).-0.682
70-80 10 75 1 - 10 10 29 22.3
Pearson's
80-90 25 85 0 0 0 54 find out the Karl
following series,
90-100 30 95 30 30 84 Example 6 From the
100-110 36 coefficient of skewness.
105 2 72 144120 15
110-120 50 115 11213
14
3 150 450 170
120-130 60 125
Measurement
39 64 3
A 240 960 230 (May 2001)
130-140 70 135 5 350 1750 300
Frequency
300
| 778 3510
 Measures of SKewness and Kurtosis MSK1

Solution
R e s e a r c h

Operations

and
Statistics

MSK-10
Business

cf
10-20 358 358
Solution
M e a s u r e m e n t f + F r e q u e n c y

a
20-30 2417 2775
d=x-A fd

AA 13 6 12
30-40 976 3751
40-50 129 3880
-9 9
50-60 62 3942
9 0
12 0 60-70 8 3960
0
13 6 4 4
70-80 10 3970
14 12
2
37
N=3970|
15
25 =N=25 item
fd=-5, 2fa* =37, 2f class Size of
A=13.

3970
Mean = 4 + 2 = Sizeof item
4
-5
Mean 1 3 + = 1 3 - 0 . 2 = 1 2 . 8
992.5th item
25 2 , class 2 0 - 3 0

2a (f
=4+
"25 =48-0.04 f=2417, i=10
L=20, cf=358,
992.5-35810
= 1.44 =12
Q20+ 2417
mode =
12
By inspection method,
Karl Pearson's coefficient of skewness =20+ 634.10
2417
Mean -Mode 12.8-12 20+2.625

1.2 22.625
0.8
=0.6667
1.2 item

Calculate Bowley's coefficient of skewness for the following

Median =Sizeof
Example 7
distribution. tem
Size of2
10-2020-30 30-40 40-50 s0-60 60-70 70-80|
1985th item
Size of
S358 | 2417 976 129 62 18 1 0 =

Median class =20-30

(Oct. 2002)
 Measures of Skewness and
Research
Kurtosis M
Operations
Solution
and
Business
Statistics

No. of children per family

S K l 2

SNo. of families
- l + 2 x i 0
c
7
1985-35810
10 17
20+2417
2 16 33
16210=20+6.731
3
-
202417 25 58 Median
26.731 4 18 76
5 11 87
8-Sizeof3
8
Size of3x
992.5th item 6 95

2977.5th item Bowley's coefficient of skewness is given by

2 ,class-30-40
s. +-2Median
Sk
-
N = 95

N+1.
L 30, cf=2775,f=976,
i=10
Size of item

2971.5-271x10
= Sizeof 96
30+ 976 item =24th item =2
4
= 30+ x10
976
Sizeof3 item
30+2.075

32.075 =Size of 3«24th item

skewness
Bowley's coeficient of 72nd item
+0-2Median
4
Q-
32.075+22.625-2(26.731) (N+I item
Median =Sizeof
32.075-22.625
96 th
54.1-53.4622300.131
9.45 9.45
Size of item

Example 8 Find Bowley's coefficiet of skewness for the following = 48th item
frequency distribution.
3
No. of children
per family 2 6
S
4+2-20)=0
4-2
No. offamilies7|10 16 25 18 11 8
 Measures of Skewness and Kurtosis MSK-15
Research

Operations

Business
Statistics
and

skewness
rom
om the lowing Size of3 item
MSK-14
coefficient of
Bowley's

3
Calculate

9 =
Exam
ample
30-40 40-50 item =112.5th item
data:
0-10
10-2020-30 10
Marks
25
20
15 2, class=50-60
10
ons
No. ofpers
Marks
s0-60 60-70| 70-80
25 10
L+3-.f-xi
4

No. ofpersons 35
l12.5-80
50+ x10=50+9.29= 59.29
Solution
No. of persons 35
r Marks f+

Marks
10 Median =Size of item
10
0-10
35
10-20
25 Size of 150) item
55
20-30 20
15
70 75th item
30-40 Median
80 Median class=40-50
10
4050 115
50-60 35
25 140
60-70 Median L+ -xi
10 150 f
70-80
N=150
75-7010=45
10
Size of item by
of skewness is given
Bowley's coefficient
S - + - 2 M e d i a n

Size of item
Q-
= 37.5th item 59.29+21.25-2(45)
59.29-21.25
class =20-30
-9.46

-c.f 38.04
=L+ -xi = -0.249

37.5-35
20+ x10=20+1.25 21.25
20
 Measures of Skewness
Research and Kurtosis ISK-11
MS
Operations
and

0,class-30-40
Statistics
Business
MSK-16 from the
coefficient of s k e w n e s s data
s

Caculate
Bowley's
Evample 10
given below
10 20 30 40 50 60 70
Profts (Rs. in lakhs) Less than 50 | 56| 59 60 30+-40
8 20 40 10 x10=30+5= 35
No. ofcompanies

Solution to calculate the actal Median =Size of| item

we have

Since cumulative frequency is given,

frequencies
of companies item
Iprofits fNo.
30th item
0-10.
Median class 20-30
121 20
O 20
20-30 20
103
40

50
Median

( Median Lt4
=
N-c.f x10
30-40
6 56
40-50
S0-60 3 59 = 20+2 x10=20+5 = 25
20
60 Bowley's coefficient of skewness is given by
60-70
N=60 S t-2Median

-0
-Siz ofhen _35+15.833-2(25) 0.833
= 0.0435
35-15.833 19.167
60 item =15th item
=Size of Example 11 A frequency distrubution showed the following measures
skewness-0.4.
e,class 10-20 oflocation: Mean= 45, Median=48,and Coefficient of
deviation.
Estimate its standard
Solution
and median is given by
Coefficient of skewness based on mean
S. M e a n - M e d i a n )

10 x10= 10+5.833 = 15.833 S.D

2 0 . 44 5 - 4 8 )

9-Sie ofs item S.D.

=Size of 3 15th item=45th item

S . D X 3 )

9
-0.4 0.4
22.5
 Research
Operations
MSK-18 Business Statistics and
Measures of ewness and
distribution, the coefficient of ske. Kurtosis MSK-19
«19
Example 12 In a frequency and the lower
quartilees Annual income (Rs. in
based on quartiles is 0.6. Ifthe sum ofthe upper is lakhs)
100 and the median is 38. Find the value of the upper quartile.
No. of persons 10-12020|120 130130
Solution 50
-

14
Coefficient of skewness based on quartiles is given by Annual income (Rs. in 45
30
lakhs)
S. +9-2(Median)
SA | No. of persons
140-150
Q,-
0.6-100-2(38) Calculate Karl Pearson's
data. coefficient of skewness
- from the
following
-0 100-70-40
0.6
Weight
(lbs) 90-100|100-110 110- 120 120-130 130-
40+ No. of 140
students 2
But +Q =10040+Q + =100 18 22 21
2=100-40 60
Weight (lbs) 140- 150 150-
160 160-170 170- 180
-=30 No. of students 19
10 3 2
,40+2 =40+30 70 4 () Calculate the coefficient of skewness with the
70 data. help ofthe following

EXERCISE Age Below 10 20 30 40 50 60

No. of persons
. Calculate the Pearson's coefficient of skewness for the following data. 15
25 40 so 55 60

Midpo intx 10 15 20 25 30 35 (i) Calculate coefficient of skewness by Karl pearson's method.

Frequency28 22 34 16 10 8
Profit (Rs. in
10-20 20-30 30-40 40-50
2 Calculate coefficient of skewness by Karl Pearson's method from the lakhs)
following data. No. of 22
18 20 30
Annual income (Rs. in lakhs) 70-80 80-90| 90-100 companies
No. of Persons 12 18 35 Profit (Rs. in 50-60
Annual income (Rs. in lakhs) 100-110D lakhs)
No. of Persons No. of 10
42
companies (April 1999)
 SKewness and Kurtosis MSK-21
Research

Operations M O M E N T S

and
data i
ata by using
S t a t i s t i c s

Business
from the
following T h e
arith
hmetic mean of the various
powers of the deviations
SK20

of
skewness
a n y

ribution is called the moments

d i s t r i

dev from mean in

about mean or central moments of
coefficient

the
S C a l c u l a t e

75 9
60 75 900 tion.
h ed i s t r i b u t i o

central momen
quartiles 15 30 60 45
Above0 20 53
65 20 The th moment about mean denoted by
Marks
180 160 130
100
65 , is given by
from the foll
No. ofstudents
coeficient ofskewness following
pearson's
Calculate Karl

data. 29.5-32.5
26.5-29.5 29.5-32.5 n
23.5-26.5
20.5-23.5
399 194
193
f17 =-X =0(4 is always
r32.5-35.5 35.5-38.5 zero)
|S27
10

coefficient of
skewness.
x-X-Variance
n

For a frequency distribution,

Calcuiate the Rewley's
65 70 75 80
Weights 60 61 62 63
No.ofstudens 3 5|7 10 3
coefficient of skewness
from the following. When the actual mean X is in fractions, it is difficult to calculate
Calculate Bowley's
60 70 moments about mean by applying the above formula. In such case, we
Marks Below 10 20 30 40 50 first compute moments about an arbitrary value Acalled raw moments and
Students 9 17 29 45 60 70 then convert these moments into moment about mean.
Marks 80 90 100 The rth moment about any point A is given by
Students 78 83 85
frequency distribution, the coefficient of skewness based
on
9. In a

quartiles is 0.5. If the sum of the upper and lower quartiles is 28 and
the median is 11. Find the values of lower and upper quartiles. For a frequency distribution,
10. You are giventhat S=0.8,Mean =40, and Mode 36. Find the value 2-4)
of standard deviation.

ANSWERS
fd
1. 0.553 2 -0.266 3. 0.812 or a
4 -0.053 5. ()-0.113,(ii)-0.159 6. 0.068 where d-
7 8. 0.014
10. 5 9 20,8 c
 Research
Operations
Statistics and
MSK-22 Business
about mean:
A into moments
COnversion ofmoments about
To obtain moments aboyt mean, we elationship
apply the following rela
m e c n
4 =
-' 0 =
Vo'nte.
H4-(4)
4, - 3 4 +2(4)"
A, - 4 4 +64(4)-3(4)'
Skewness
A measure of skewness is obtained by making use of the second and
third
moments about mean and is denoted by B.
It is defined as
B, is always positive because we are squaring 4 and
variance is always h being the
positive. Thus B, cannot tell us about the direction of
skewness.
This drawback is removed if
we calculate
root of 7, defined as the square
, (i.e.)
If 7, =0, the distribution is
distribution is positively skewed, symmetrical. If
and if y, <0, then Y,>0, then the
negatively skewed. the distribution5
Kurtosis
Kurtosis is a measure which
frequency curve of a studies about the flatness or
The measure distribution. peakness of tne
of kurtosis is denoted by B, and is defined as
Y2 obtained
d from
B,H
defined by B, is also used as a
measure off kurtosis and is
Y B-3
 Measures of Skewness and Kurtosis MSKz3
or a normal curve, Y, =0 (i.e.)B, = 3. Then the curve is calleed
For a norma
nebokurtic.
en Y2 is
Wher positive (i.e.) B, >3, the curve is more peaked than the
al curve and is called leptokurtic.
When is negative (i.e.) B, <3, the curve is less peaked than the
al curve and is called platykurtic.
norm
Lepto kurtic
Meso kurtic
Platy kurtic
Fig. 5.1
WORKED EXAMPLES
Example Compute the first four central moments for the following
data.
8,10,11, 12, 14.
Solution
F -
a-¥|a-Ry|-F
3 9 -27 81
- I
10 1
11 0 0 0 0
1
12
9 27 81
14 56 164
55 20
n
n
n
 Research
MSK 24 Business Statistics and Operations

H x- =112
r-X)l64 =32.8
n

Example 2 The first fourmoments of a distribution about the value 4

Aof
of
a variable are-1.5, 17,-30 and 108. Find the central moments, B, and R
Solution Moments about the value 4:
-1.5,4 =17,4=-30,4 = 108
Moments about mean:
4-=0
4-(40
17-(-1.5 =14.75
4-3 K+2(4
-30-3(17-1.5)+2(-1.5
-30+76.5-6.75= 39.75
4-445 H+644 (4{ -3(4)
= 108-180+229.5-15.1875
=108-4(-30X-1.5)+6(17X-1.5 -3(1.5)
142.3125

A 0975
(14.75
0.4926
A -142.3125
H (14.75
0.6543
Example 3 The first four central
Comment on the moments are 0, 2.5,
skewness and kurtosis of the 0.7, and 18.7
Solution distribution.
=0, 4=2.5, j
=0,7, 4, =18.75
 Measures of Skewness and Kurtosls MSK-25

skewness

=0.7) = 0.031
(2.5
Since, B#0, the distribution is not symmetrical.
Kurtosis

B, 2.5
since B, = 3, the distribution is mesokurtic.

ample 4 The first four moments ofa distribution about the value 5 are
Exam

.20. 40 and 50. Obtain the mean, variance, B, and B,

Solution

Given: 4=2, 4 = 20,4 =40, 4 = 50

Mean= A+H
=5+2 7
Variance=4, = 4 - ( 4 )

20-(2)=16

4h 4-34+2(4
40-3(2 20)+2(2)
=-64

H444 +6(4°'42-3(4)*
=50-4(2(20)+6(2)* (20)-3(2
322

B- 4.
(16
322
= 1.26
 Statistics and Operations
Research
Measures 0 Kewness
and
MSK-26 Business
for the folla. Example 6 Calculate the first fourr Kurtosis MSK:27
Calculate the first four
central
moments
wing data.
ur
moments about mean
Example for the follov
llowing
data.
35 45 Ss 65 75
s
Solution
3
|11|21 43 |8325 959
Solution

a - )
dz-65
0
35
fa' sa
-2 -3
45 9 -27
6 81
55 11 12 24
339 48
65 21 -11
0
75 43 1 0
43 13
S31ST 85 32 2 64
43 43
12 9 128 256 512
2 95
27 81 243 729
120 | 114 284
Moments about the value 65
480 1424
Sr-Ry r-Ry|Sr-
16
= i=x10=9.5
120
4

3 3
H x
2
i =x100
120
=236.67
0 0
480
3 = x? = x1000 =4000
8 6
120

14 0 38 u=2 fay14210000=l 18666.67

120
-X =4
15 Moments about mean:
-=0
0 4 -(4) =236.67-(9.5) =146.42
f
4 -34s +2(4
-1030.345
==0.933
5
=
4000-3(236.67)(9.5)+2(9.5)} =

H =H-4,4'+64;(4/)-3(4)"
= 0
=| 18666.67-4(4000)(9.5) +
6(236.67X9.5)* -3(9.5
I5
=l18666.67-152000+128156.81-24435.188
A=-XY_38 2.533 70388.29
wSK-28 Business
Statistics and Operations
Research
Measures ness and Kurtosis
following distrit.. MSK-29
Evample 7
Calculate the first four
moments ofthe stribution
and hence
find and P,.
about the mean

Solution
H, -X_216
26
s 8 2 8 56 70
56|28 8 Skewness:

Sx-)
-4 -4
4 =0, the
0 Since distribution is
8
2
-24
56 Kurtosis:
symmetrical.
28 56
168
56
3 S6
0 0
70 280
280 56
56
168 2 56 2.75
28 since B, <3, the distribution is platykurtic.
S6 24
8
4
8 Exarmple 8 Analyze the following distribution
0 by the method of
1 256 1024 moments.

Sa-F? - Sr- 2 4 6 8 10 12 14
64 256 f4 11 |18 27 20 16 8
16
648 Solution
72 -216
112 -224 448
56 -56 56 2
0 0
0 -12 -I08 324
56 56 56 4 11 -22 44 83 176
448 18 -18 8 -13 18
112 224
27 0 0
2 216 648
20 20 20 20
256 10 20
16 64 64 128 256
12 16
216 648
512 0 2816
14 8 24 72 130 1442
104 4 254
F 1024 4 Moments about the value 8:
256

Ef-X)0 KRAMES
SWAR
2 Jd x4=9.
fa-X S12 4 104 LIR LI
2f 256 Ar
 easures of Ske
Research
skewness and Kurtosis
Statistics
and
Operations
Solution
MSK-31
Business
SK-30

A-2Jdx= 8=10 Variable

Midpoint
d 22.s
2 Jd
u = 2 J x t =x16=221.85 0-5
fa
5-10
5 7.5 -3 32
- 128 512
12.5 5 45
Moments about mean:
10-15
-2 -135 405
13 17.5 -14 28
1 5 - 2 0

56 112
0 -13 13
=9.56 21 22.5 -13 13
=9.77-(0.46) 20-25
0
-(4 0
4
25-30 16 27.5 0 0 0
4 4-34(4)+2(4"
16 16
-3.29 32.5 2
16 16
+2(0.46)=
3 0 - 3 5

10-3(9.77K0.46) 16 32
=
35-40 3 37.5 64 128
=4-44 K+64,(4-3(4 9 27 81 243
221.85-4(10)X0.46)+6(9.77)(0.46)* -3(0.46) 75 -9 193 171
221.85-18.4+12.40399-0.1343 Moments about the value 22.5:
1429

= 215.7197 fd
Skewness:

B
-3.29)
(9.56)
10.82410.0124
873.7228 x 64.33
symmetrical. But 4, <0, thusthe
B>0, the distribution not
is
Since
distribution is negatively skewed.
1429
Kurtosis: u=2xt =x625=1908.33
215.719-2.3603
(9.56)
Moments about mean:

= 0
Since B, <3, the distribution is platykurtic.
about mean for the following
Example 9 Calculate the first four moments 64.33-(0.6) = 63.97 I S7 y D 3

data:
h-34+2(49
20-25
Variable 0-5 S-10 10-15 15-20 285-3(0.6)%64.33)++2(0.6) =169.638
13 21
Frequency2 5 7 A -44H+64,(4*-34
Variable 25-30 30-35 35-40 -30.6) =11362.895
=11908.33-4(285 0.6)+6(64.33)M0.6)
Frequency 168 3
 Operations Research
MSK 32 Business Statistics and

Evample l0 Calculate the first four moments of the lollowing distribim

the moments about the mean
AN.
about the value 40.45 and hence find
Class interval 30.0-32.9 33.0-35.9 36.0-38.9|39.0-41.9
2 4 26 47
Frequency
45.047.9
Classinterval 42.0-44.9
15 6
Frequeney
Solution

Midpoint
Class interval d fd

2 -6 18-54| 162
30.0-32.931.45
33.0-35.9 34.45 4 -8 16 -32 64
36.0-38.9 37.45 26 26 26 26 26
39.0-41.9 40.45 47 0 0
42.0-44.9 43.45 15 15 15 15 5
45.0-47.9 46.45 6 2 12 24
4896
100 - 13 99-49 363
Moments about 40.45

=9i=-3,2
100

= x =x9
100.
=8.91
H x? =x27=-13.23
100

x-363 x81 294.03

f
Moments about mean:
=0

4 4-(4 =8.91-(-0.39 =8.76

4h4-34%+2(4,"
=-13.23-3(-0.39)(8.91)+2(-0.39)
-2.928
 Measures of Skewness and Kurtosis MSK-33

H, =H-44H+64$(4)*-3(4)*
294.03-4(-13.23X-0.39)+6(8.91K-0.39)-3(-0.39)
281.45le

Example ll
mole 11 Calculate the value of B, and A, from the following data
and interpret them.

Wages (Rs.hundreds)| 40-50 50-60 60-70 70-80 80-90

No ofworkers 10 25 30 23 12
Solution

Midpont
Wages d=5 fd fa
10

40-50 45 10 -2 -20 40 -80 160

55 25 - 1
s0-60 -25 25-25 25
60-70 65 30 0 0
70-80 75 23 23 23 23 3

80-90 85 12 24 48 96 192

100 | 2.13614 400

-Xix10-02
100
136 136
100

Ya -xi' =x1000= 140

100 100

u =x=x10000=40000
100

=136-(0.2) =135.96
4 4-(4
4 -34H+2(4
=140-3(0.2X136)+2(0.2)
= 58.416
 Research
Statistics and Operations
MSK-34 Business

, =-44'u+64,(4) -3(4)*
(136)-3(0.2)*
= 40000-4(0.2X140) +6(0.2)'
39920.64

Skewness
(135.96 0.0014
(58.416)=0.0014
Since B, 0 , thedistribution is not symmetrical. But >0, hence
ence
the distribution is skewed.
positively
Kurtosis:
39920.64 2.16
( 1 3 5 . 9 6 )2 . 1 6

A-
Since B2 <3, the distribution is platykurtic.
Example12 Calculate Skewness and Kurtosis for the following
distribution.
Marks 0-10 10-20 20-30 30-40 40-50 50-60
Frequency | 5 20 15 45 10 5
Solution

Marks MidpointS d -35Ja T

10
0-10
-15 45 -135405
10-20 15 20 -2 40 80 160 320
20-30 25 15
15 - 15 15
30-40 35 45
- 15
0
40-50 45 10
10 10 10
50-60 55 5 2 10 20 40 80
100
50 170-260 830

xix0
100x10= -5

10X100
100 =170
 Measures of Skewness and Kurtosls MSK 35

-260
X/' 100 x1000-2600
w 830 83000
100
Moments about mean;

- 0
=
4 -(4) 170-(-5)
= 145

4, -34+2(4
-2600-3(170)X-5)+2(-5)
=-300

,H-4u+64,04*-34,)"
83000-4(-2600X-5)+6(170)(-5)*-3(-5)*
54625
Skewness:

A- 300) 90000 0.0295

3048625
(145(145)
<0 thus the
Since 0 the distribution is not symmetrical. But 4,
distribution is negatively skewed.

Kurtosis:

B 2.598
platykurtic.
Since B, < 0 the distribution is
EXERCISE
following data.
moments about 10 for the
Find the first four
-

8,10,11, 12, 14. 22, I17,

5 are equal to-4,
moments about the value
f thefirst four
moments about mean.
S60, determine the about.r =4 are 1, 4, 10 and 45.
distribution of the
the nature
3 moments of a on
comment
he first four mean and
moments,
central
ind the
distribution.
 MSK-36 Business Statistics and Operations Research

The first four central moments about the distribution are

2,6, and
100. Find and
B, P,
5. Calculate the first four moments about the mean and the value o
for the following data.
of B,

0 2 4s67
S510 15 20 25 20 1 10
Compute the coefficient of skewness and kurtosis based on the
following data. the

4.5 14.5 24.5 34.5 44.5 54.5 64.5

s 5 12 22 179 4
74.5 84.5 94.5
f|3
7. Examine the nature of distribution for the
moments.
following by the method of

s26 13 25 30 2 9 5 2
& From the
following data calculate moments (i) about the value 25
(ii) about mean and

Variabe 0-10 10-20 20-30 30-40

Frequency 1
3 4
Calculate the first four moments and
B,from the following data. find the value of and
B,
Marks 0-20 20-40 40-60 60-80 80-100
No. of students 8 12 20
10. Calculate the skewness and
6
kurtosis for the
Profits (Rs.
given data.
lakhs)10-20 20-30 30-40 40-50 50-60
50-60
No of companies 18 20 30
22 1010
 Measures of Skewness and Kurtosi MSK37
alculate the
first four central moments, 9, and B, for the following
d
etribution and interpret the results.
Age 25-3030-35 35-40 40-45 45-50
Frequency 2 8 18 27 25
Age 50-55 55-60 60-65
Frequency | 16 7 2
ANSWERS
LH=1, H =4.8, u = 0.8, u = 57.8
2 4 0 ,4 =6, 4, =19, 4, =32

3 0, 4, =3, 4 =0, = 26, B, =0, B, = 2.89, mean =5

= =
4 B 0.667, B, 2.78

5. =0, 4=4, 4=0,

4, =37.6, B =0, B, =2.35
6. H 0, 4 -2.83, 4 =3.38, H, =30.295, , =0.504, B, =3.782,
Leptokurtic
7. B, =0.03, B, =3
8. =-3, =90, =-900, 4 =21000
= 0, 2 =81, 4, =-144, 4, =14817
9. B 0.0486, B, = 2.563
10. = =
B 0.00013, B, 2.047
11. = =
0, u, 2.238, u, =0.1955, H, =12.966
B 0.0034, B, =2.59